Stroke is an important cause of death and disability among children, for whom there is no proven therapy except blood transfusion for sickle cell anemia. Research leading to improved treatment of stroke has largely excluded children. Progress in understanding pathophysiology and improved treatment of childhood stroke requires multicenter prospective cohort studies to define reliable and valid outcome predictors. Determinants of outcome in childhood stroke are unknown. Clinical stroke severity measured by the NIH Stroke Scale (NIHSS) is a robust outcome predictor in adults, but has not been studied in children. The purpose of this project is to evaluate the reliability, validity and predictive utility of a pediatric adaptation of the NIHSS, (PedNIHSS). There are three specific aims, to assess: 1) the relationship of PedNIHSS to acute infarct volume and location measured on brain MRI; 2) predictive utility of PedNIHSS score relative to 3- and 12-month neurologic and functional outcomes; 3) inter-rater reliability and predictive sensitivity and specificity of the PedNIHSS. In a consecutive prospective cohort study, 8 pediatric tertiary care hospitals in the International Pediatric Stroke Study network will enroll 120 subjects over 18 months, to include all children age 2-18 years with acute arterial ischemic stroke, excluding patients with acute head injury, cardiac arrest, brain tumor or meningitis. PedNIHSS will be recorded daily for 7 days, and correlated with infarct volume on T2-weighted MRI obtained on admission and at 5-7 days after symptom onset. Predictive utility of PedNIHSS will be assessed by multivariate regression models for 3- and 12-month outcomes which will include measures of neurologic impairment using the Pediatric Stroke Outcome Measure, disability using the Pediatric Evaluation of Disability Inventory, and quality of life using the PedsQL. Interrater reliability will be measured by comparing PedNIHSS scores given by a primary examiner with that of a second examiner observing the examination. The results of this study will provide new information about how initial clinical stroke severity is related to size and location of the stroke, its importance in predicting outcome in children, and the influence of age, gender, and stroke risk factor on the prediction of outcome. These observations, and the availability of a simple valid clinical stroke scale for children, will be important for planning and conducting future clinical trials and outcome studies for childhood stroke. [unreadable] [unreadable]